Skill UI

Provides programmatic access to the CZ CELLxGENE Census (61M+ cells) for querying standardized single-cell expression, metadata, embeddings, and stats across tissues and diseases, supporting scanpy and ML workflows.

Molfeat is a Python toolkit that converts SMILES/RDKit molecules into embeddings via 100+ calculators, transformers, and pretrained models for QSAR, screening, similarity search, and deep learning pipelines.

Use UMAP for fast nonlinear manifold reduction to produce embeddings that keep local and global structure, making it ideal for visualization, clustering preprocessing, supervised/parametric workflows, or high-dimensional feature engineering.

Umap provides uniform manifold approximation for dimension reduction and visualization, helping data scientists explore high-dimensional embeddings through interpretable 2D/3D layouts.

This skill automatically generates an interactive SQL Notebook designed to validate changes in dbt models or snapshots. It processes changes identified from a GitHub PR or local repository, generating comparison queries that compare 'before' (production) and 'after' (development) data states. It is essential for embedding data observability and quality checks into CI/CD pipelines.

Provides programmatic and standardized access to massive, versioned datasets of single-cell and spatial transcriptomics from CZ CELLxGENE. This skill allows users to query millions of cells, analyze gene expression, and access pre-calculated embeddings across various tissues, diseases, and species without downloading entire datasets. Essential for large-scale bioinformatics analysis, machine learning model training, and comprehensive cross-dataset comparison.

UMAP (Uniform Manifold Approximation and Projection) is a powerful technique for non-linear dimensionality reduction. This skill enables generating scalable and structure-preserving embeddings for visualization (2D/3D), clustering preprocessing, and feature engineering. It supports supervised and semi-supervised workflows, allowing users to guide the embedding process with labeled data while maintaining local and global data structure integrity.

A comprehensive guide to diagnosing and enhancing search relevance within Qdrant. This skill addresses reported issues such as low precision, low recall, or irrelevant matches. It guides users through determining if the problem lies in the embedding model, Qdrant configuration, or the query strategy. Techniques covered include optimizing chunk splitting, tuning HNSW parameters, implementing hybrid search, and utilizing reranking for superior result quality.

Provides a systematic guide for diagnosing and improving poor search quality in vector databases like Qdrant. This guide covers troubleshooting issues such as low recall, when approximate search degrades, selecting optimal embedding models, and correctly tuning HNSW parameters, quantization, and filtering strategies. Essential for optimizing RAG and semantic search pipelines.

A comprehensive guide and workflow for embedding R scripts, primarily using the ggplot2 library, into Power BI Reports (PBIR). This tool enables the creation of high-quality, static, publication-ready statistical visualizations directly within the report canvas by executing R code on the defined dataset. It covers the full lifecycle: adding the visual, writing the script, injecting the code, and validating the output.

A spectral vector search tool that enhances nearest-neighbor search with graph Laplacian features, capturing both semantic similarity and structural roles in embedding spaces.

This tool ingests raw market data (OHLCV) for a specified symbol. It normalizes the data into relative OHLCV vectors, converts these vectors into high-dimensional embeddings, and then stores and indexes them using HNSW. This process prepares the data for fast, similarity-based pattern detection and advanced financial time series analysis.